Protein–Ligand Docking Using Hamiltonian Replica Exchange Simulations with Soft Core Potentials

Abstract

Molecular dynamics (MD) simulations in explicit solvent allow studying receptor-ligand binding processes including full flexibility of the binding partners and an explicit inclusion of solvation effects. However, in MD simulations, the search for an optimal ligand-receptor complex geometry is frequently trapped in locally stable non-native binding geometries. A Hamiltonian replica-exchange (H-REMD)-based protocol has been designed to enhance the sampling of putative ligand-receptor complexes. It is based on softening nonbonded ligand-receptor interactions along the replicas and one reference replica under the control of the original force field. The efficiency of the method has been evaluated on two receptor-ligand systems and one protein-peptide complex. Starting from misplaced initial docking geometries, the H-REMD method reached in each case the known binding geometry significantly faster than a standard MD simulation. The approach could also be useful to identify and evaluate alternative binding geometries in a given binding region with small relative differences in binding free energy.

Fully Blind Peptide-Protein Docking with pepATTRACT

“…Recently a molecular dynamics-based approach was developed in our group, which employs Hamiltonian replica exchange simulations with a variation of soft-core potentials along the replicas. This method showed promising results with respect to local peptide-protein docking (Luitz and Zacharias, 2014).…”

“…In other words, such an approach should predict both the peptide binding site (global search of the protein surface) and the bound peptide conformation to high precision simultaneously. A number of peptide-protein docking and binding site prediction tools have been developed to date (Antes, 2010;Bordner and Abagyan, 2006;Dagliyan et al, 2011;Donsky and Wolfson, 2011;Dundas et al, 2006;Heté nyi and van der Spoel, 2002;Lavi et al, 2013;Luitz and Zacharias, 2014;Niv and Weinstein, 2005;Petsalaki et al, 2009;Raveh et al, 2011;Rosenfeld et al, 1995;Saladin et al, 2014;Staneva and Wallin, 2009;Trabuco et al, 2012;Unal et al, 2010;Verschueren et al, 2013). Global docking and binding site prediction methods (Ben-Shimon and Eisenstein, 2010;Dagliyan et al, 2011;Dundas et al, 2006;Lavi et al, 2013;Petsalaki et al, 2009;Saladin et al, 2014;Trabuco et al, 2012;Verschueren et al, 2013) often identify the correct binding site but do not yield high-quality models for the peptide conformation (London et al, 2013b).…”

Attracting cavities for docking. Replacing the rough energy landscape of the protein by a smooth attracting landscape

“…These approaches are generally unable to cross high‐energy barriers of ligand–protein interaction. Therefore, the ligand–target complex geometry is often trapped in local energy minima corresponding to nonnative binding geometries, although it was shown recently that the global minimum can be determined using replica exchange simulations in some cases . As a consequence, simulation approaches are seldom used as stand‐alone sampling algorithms.…”

“…Therefore, the ligand-target complex geometry is often trapped in local energy minima corresponding to nonnative binding geometries, [26] although it was shown recently that the global minimum can be determined using replica exchange simulations in some cases. [27] As a consequence, simulation approaches are seldom used as standalone sampling algorithms. However, they can efficiently improve other search methods by locally refining poses suggested by MC-, EA-, or SI-based algorithms, for instance.…”

An enhanced-sampling MD-based protocol for molecular docking

“…Clearly, due to time scale restrictions, standard MD simulations are often unable to sample conformational states relevant to molecular recognition 38,52 . Therefore, several techniques have been proposed to enhance the sampling of rare conformations, including accelerated MD 53 , replica-exchange in temperature and energy spaces 54,55 , and metadynamics 56 , which generalizes methods such as conformational flooding 57 and local elevation 58 . Several groups have demonstrated the power of these methods (sometimes coupled with the use of cosolvents) in improving the performance of docking and virtual screening [59][60][61][62][63][64][65][66] .…”